Coreutils POSIX changes not documented in release notes

[Paul W. Frields]

On Thu, 2006-03-09 at 08:25 -0500, Bill Rugolsky Jr. wrote:
> On Thu, Mar 09, 2006 at 06:42:17PM +0530, Rahul Sundaram wrote:
> > >You can disagree all you want, but this is a change from FC4 to FC5:
> > >
> > >fc4% sort +2 < /var/log/messages | head -1
> > >Mar  7 00:27:35 ti63 smartd[2187]: Device: /dev/sda, Temperature changed 
> > >-2 Celsius to 35 Celsius since last report
> > >
> > >rawhide% sort +2 < /var/log/messages | head -1
> > >sort: open failed: +2: No such file or directory
> > >
> > > 
> > >
> > This seems unrelated to the man page entry you quoted. Can you find more 
> > information on this and provide us the content?
> 
> I don't mean to be an ass, but do you read before you mouth off?  Judging by
> the fedora-* list traffic, no.  Let's try again:

No need to be rude, Rahul wasn't.  He was matter of factly stating a
request, not questioning your intelligence.

>        1003.1-2001.  For example, if you have a newer system but are running
>        software that assumes an older version of POSIX and uses `sort +1' or
>        `tail +10', you can work around any compatibility problems by setting
>        `_POSIX2_VERSION=199209' in your environment.
> 
> Sorry, I'm cranky.  Now I have to go put a wrapper around coreutils and
> log old-style arguments.
> 
> Environment variable tests are completely broken.  That was already known
> in the 1970's ...  but I gave up on POSIX when I was a reviewer for the
> pthreads drafts ...

The change in this behavior in GNU sort(1) was noted when the POSIX
standard changed back in 2001 or so.  I believe there was a lot of talk
around that time about the new behavior.  It was known at that time that
scripts using the [soon to be] deprecated "sort +N" behavior would
[eventually] need to change under GNU coreutils.  Looks like they
finally broke the compatibility to make the change complete, at least in
GNU sort(1).  Certainly we can note *that* in the release notes, and if
you're able to track down other specific breakages between FC4 and FC5,
those would be good candidates for relnotes as well.  Thanks for your
efforts.

-- 
Paul W. Frields, RHCE                          http://paul.frields.org/
  gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233  5906 ACDB C937 BD11 3717
 Fedora Documentation Project: http://fedora.redhat.com/projects/docs/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 191 bytes
Desc: This is a digitally signed message part
URL: <http://listman.redhat.com/archives/fedora-docs-list/attachments/20060309/dc1fd2ba/attachment.sig>